How do I remove vocals from a song?

Load the song into PumpyDumpy2AudioMaster and run stem separation. It uses Demucs to split the track into vocals, drums, bass, and other instruments, then you mute or drop the vocal stem to keep the instrumental. The whole process runs locally on your PC, so there is no upload and no per-song limit.

Can I make an acapella (vocals only) too?

Yes. The same stem separation that gives you an instrumental also isolates the vocal on its own, so you can export an acapella for remixes, mashups, or sampling. You get both sides of the split from a single pass.

Does the vocal remover upload my song to a server?

No. PumpyDumpy2AudioMaster is a fully offline desktop app. The Demucs model is bundled into the installer, so your song is processed entirely on your machine - no upload, no cloud queue, and no waiting in line behind other users. This also keeps unreleased or copyrighted material private.

What quality can I expect from the instrumental and acapella?

Demucs is one of the strongest open-source separation models, so results are clean on most modern productions. Very dense mixes or heavy effects can leave faint bleed between stems, which is normal for any separation tool. Because it runs locally, you can re-run with different settings as many times as you like to get the best split.

Vocal Remover: Remove Vocals or Make an Acapella, Fully Offline

Q: Is the vocal remover free?

Yes. PumpyDumpy2AudioMaster is free to download and use, with a light watermark on exported audio. Pro removes the watermark and grants a commercial-use license for $20/month or $200/year, and the same subscription also unlocks PumpyDumpy2Visual.

Q: Can I remove vocals from any song or only AI tracks?

Any song. PumpyDumpy2AudioMaster works on AI tracks from Suno or Udio, your own recordings, and any audio file you have the right to use. The separation does not care how the song was made - it works from the audio itself.

Split any song into an instrumental and an acapella on your own PC - no upload, no per-song limit, no waiting in a cloud queue.

Most vocal removers are websites: you upload your song, wait in a queue, and hope the free tier lets you download the result. That is slow, it puts your audio on someone else's server, and it caps how many tracks you can process. PumpyDumpy2AudioMaster (PD2AM) takes the opposite approach - it is an offline Windows desktop app that removes vocals locally, so nothing is uploaded and there is no limit on how many songs you run.

Under the hood it uses Demucs, one of the best open-source source-separation models, to split a song into clean stems. From there you can keep the instrumental, isolate the acapella, or export every stem for a remix.

Instrumental, Acapella, or Full Stems

A vocal remover is really a stem splitter, and PD2AM gives you everything the split produces:

Instrumental: drop the vocal stem to get a clean backing track for karaoke, covers, or practice.
Acapella: keep only the vocal for remixes, mashups, and sampling.
Full stems: export vocals, drums, bass, and other instruments separately for full remix control.

You get all of this from a single separation pass, so one run covers every use case.

Offline and Private by Design

Because the Demucs model is bundled into the installer, your song never leaves your computer. There is no upload, no cloud queue, and no account that stores your files. That matters when you are working with unreleased material, client tracks, or anything you simply do not want sitting on a third-party server. It is also faster in practice: no upload time, no download wait, and no daily cap.

How Vocal Removal Works

Old-school vocal removers tried to cancel the centre channel, which mangled the mix and only worked on simple stereo songs. Modern separation is different: Demucs is a trained model that listens to the audio and rebuilds each instrument as its own track. That is why the instrumental and acapella sound natural instead of hollow. Load a file, run the split, preview the stems, and export the ones you need - it works the same whether the song is a polished commercial release or an AI track.

Works on Any Song, Not Just AI

PD2AM is built around AI music, but the vocal remover does not care how a song was made. It works on AI tracks from Suno and Udio, your own recordings, and any audio file you have the right to use. If you can load it, you can split it.

More Than a Remover: Repair, Master, and Deliver

Removing vocals is one tool in a full audio pipeline. Once you have your stems, PD2AM can repair the vocal, denoise with DeepFilterNet, master the mix to streaming loudness, transcribe lyrics into an .lrc with Whisper, and export a delivery pack. It also bridges to its companion app PumpyDumpy2Visual for turning a track into a music video.

Pricing

PumpyDumpy2AudioMaster is free to download and use, with a light watermark on exported audio. Pro removes the watermark and grants a commercial-use license for $20/month or $200/year. The same Pro subscription also unlocks PumpyDumpy2Visual, so you get the audio toolkit and the music-video app for one price. Already a PumpyDumpy2Visual Pro subscriber? Your existing key activates PD2AM too.

Download PD2AM Free for Windows See PumpyDumpy2AudioMaster